AgentToolHub
Skill

mcp-server-development

mcp-server-development — a Claude Code skill from akiojin/llmlb

Features

  • ✓**Complete Request/Response Logging**: Captures full request bodies,
  • ✓**Automatic Retention**: Keeps history for 7 days with automatic cleanup
  • ✓**Web Dashboard**: View, filter, and search request history through the
  • ✓**Export Capabilities**: Export history as CSV
  • ✓**Filtering Options**: Filter by model, runtime, status, and time range
  • ✓Linux/macOS: `~/.llmlb/lb.db`
  • ✓Windows: `%USERPROFILE%\.llmlb\lb.db`

Installation Guide

Prerequisites

  • Linux/macOS/Windows x64 (GPU recommended)

  • Rust toolchain (stable) and cargo

  • Docker (optional)

  • CUDA Driver (for NVIDIA GPU) - see [CUDA Setup](#cuda-setup-nvidia-gpu)

    • Pre-built Binaries

    Download platform-specific binaries from GitHub Releases.

    | Platform | Files |
    |----------|-------|
    | Linux x86_64 | llmlb-linux-x86_64.tar.gz |
    | macOS ARM64 (Apple Silicon) | llmlb-macos-arm64.tar.gz, llmlb-macos-arm64.pkg |
    | macOS x86_64 (Intel) | llmlb-macos-x86_64.tar.gz, llmlb-macos-x86_64.pkg |
    | Windows x86_64 | llmlb-windows-x86_64.zip, llmlb-windows-x86_64-setup.exe |

    #### macOS Notes

    The macOS .pkg installers are not code-signed, so you'll see a security warning on first run.

    To install:

  • Right-click the .pkg file in Finder → Select "Open"

  • Click "Open" to proceed

    • Or remove the quarantine attribute via Terminal:

    bash
    sudo xattr -d com.apple.quarantine llmlb-macos-*.pkg

    CUDA Setup (NVIDIA GPU)

    For NVIDIA GPU acceleration, you need:

    | Component | Build Environment | Runtime Environment |
    |-----------|-------------------|---------------------|
    | CUDA Driver | Required | Required |
    | CUDA Toolkit | Required (for nvcc) | Not required |

    #### Installing CUDA Driver

    The CUDA Driver is typically installed with NVIDIA graphics drivers.

    bash
    # Verify driver installation
nvidia-smi

    If nvidia-smi shows your GPU, the driver is installed.

    #### Installing CUDA Toolkit (Build Environment Only)

    Required only for building the runtime with CUDA support (BUILD_WITH_CUDA=ON).

    Windows:

  • Download from CUDA Toolkit Downloads

  • Select: Windows → x86_64 → 11 → exe (local)

  • Run the installer (Express installation recommended)

  • Verify: Open new terminal and run nvcc --version

    • Linux (Ubuntu/Debian):

    bash
    # Add NVIDIA package repository
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt update

# Install CUDA Toolkit
sudo apt install cuda-toolkit-12-4

# Add to PATH (add to ~/.bashrc)
export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

# Verify
nvcc --version

    Note: Runtime environments (runtimes running pre-built binaries) only need the CUDA
    Driver, not the full Toolkit.

    1) Build from Rust source (Recommended)


    bash
    git clone https://github.com/akiojin/llmlb.git
cd llmlb
make quality-checks   # fmt/clippy/test/markdownlint
cargo build -p llmlb --release

    Artifact: target/release/llmlb

    2) Run with Docker


    bash
    docker build -t llmlb:latest .
docker run --rm -p 32768:32768 --gpus all \
  -e OPENAI_API_KEY=... \
  llmlb:latest

    If not using GPU, remove --gpus all or set CUDA_VISIBLE_DEVICES="".

    3) External Endpoint Runtime


    Set up the inference server you want to register separately. llmlb works with LM Studio, Ollama,
    vLLM, xLLM, and other OpenAI-compatible endpoints.

    Requirements

  • LLM Load Balancer: Rust toolchain (stable)

  • External endpoint runtime: Requirements depend on the server you register

    • Reviews